quant-ph digest — 2026-05-10

Generated 2026-05-10 · 73 entries scored · 7 relevant

Scored against Yuan's research programme (Y1–Y6):

Source

arXiv listing: https://arxiv.org/list/quant-ph/new (55 new + 18 cross = 73 entries; announce cycle Friday 8 May 2026)

Coverage: all 73 entries scored. 7 relevant (score ≥ 1); 66 SKIP (score 0, omitted).

Scoring rubric

0–10 on method/scope/conclusion overlap — max wins. HIGH 8–10 · MED 5–7 · LOW 1–4 · SKIP 0.

Highly relevant (score 8–10) — 0 papers

No HIGH-scoring papers in today's announce cycle. No deep-analysis pass run.

Moderately relevant (score 5–7) — 2 papers

Quantum-enhanced Large Language Models on Quantum Hardware via Cayley Unitary Adapters

Large language models (LLMs) have transformed artificial intelligence, yet classical architectures impose a fundamental constraint: every trainable parameter demands classical memory that scales unfavourably with model size. Quantum computing offers a qualitatively different pathway, but practical demonstrations on real hardware have remained elusive for models of practical relevance. Here we show that Cayley-parameterised unitary adapters -- quantum circuit blocks inserted into the frozen projection layers of pre-trained LLMs and executed on a 156-qubit IBM Quantum System Two superconducting processor -- improve the perplexity of Llama 3.1 8B, an 8-billion-parameter model in widespread use, by 1.4% with only 6,000 additional parameters and end-to-end inference validated on real Quantum Processing Unit (QPU).

Architecture Shape Governs QNN Trainability: Jacobian Null Space Growth and Parameter Efficiency

Variational quantum circuits with angle encoding implement truncated Fourier series, and architectures arranging $N$ qubits with $L$ encoding layers each -- sharing encoding budget $E = NL$ -- generate identical frequency spectra, identical frequency redundancy, and require the same minimum parameter count for coefficient control. Despite this equivalence, trainability varies substantially with architecture shape $(N,L)$ at fixed $E$. We identify structural rank deficiency of the coefficient matching Jacobian $J$ as the mechanism responsible. For serial single-qubit architectures, we prove $\mathrm{rank}(J) \leq 2L+1$ regardless of parameter count $P$, with $\dim(\ker J) \geq P-(2L+1)$ growing without bound — a phenomenon we term structural gradient starvation: a growing fraction of parameters become structurally decoupled from the loss as $P$ increases at fixed $L$.

Tangential (score 1–4) — 5 papers

Summary table

ScorearXiv IDShort titleOverlapsarXiv
52605.05914LLMs via Cayley unitary adapters on IBM 156-qubit System TwoY6 (scope: hardware)link
52605.05942Architecture shape governs QNN trainability (Jacobian null-space)Y1, Y3 (method)link
32605.05477Post-quantum correlations from a standard quantum walkY6 (foundations)link
32605.05479Gross-Neveu real-time dynamics on IBM superconducting (utility scale)Y6 (scope: NISQ)link
32605.06122Variationally compressing Trotter circuits for chemistryY3 (method)link
22605.06167Variational algorithm for matrix eigenvaluesY3 (method)link
22605.06224Modular wedge, Majorana, Tsirelson limit of Bell-CHSHY6 (foundations)link